Topband
[Top] [All Lists]

Re: Topband: Norton preamp

To: "Steve Flood" <flood@ixi.net>
Subject: Re: Topband: Norton preamp
From: "Rick Karlquist" <richard@karlquist.com>
Reply-to: richard@karlquist.com
Date: Sat, 20 Jan 2007 09:38:34 -0800 (PST)
List-post: <mailto:topband@contesting.com>
Steve Flood wrote:
> Has anyone tried one of these preamp designs?
> I cannot make sense of the use of the extremely hi-mu FT-50-75 cores in
> this design.
>
> http://www.kongsfjord.no/dl/Amplifiers/Ultralinear%202N5109%20And%202N3053%20Amplifiers.pdf
>
>
> Steve, KK7UV

The transformers are standard broadband receiving type transformers,
which typically use hi-mu cores.  Get a copy of "Transmission Line
Transformers" to read about this.

Additional comments:  I have been reading Microwave Journal for 35
years and remember when the referenced articles by Norton, Rhode, etc
were published.  I remember the 2N5109 and the 2N3053.  The author
claims the 2N3053 is somehow related to the 2N5109.  I don't remember
that being the case.  We used to use the 3053 as a power driver for
bigger transistors.  I probably still have a bunch of those in my
junk box.  In any event, the precursor to the 2N5109 was the 2N3866.
This was the first big transistor with low base spreading resistance.
Low base spreading resistance is necessary to get low noise figure at
high collector currents.  Another variant of these transistors is the
2N5943.

The old Anzac company (later RHG, later MA/COM, later Tyco) made a
bunch of transformer feedback amplifiers, such as AM-123.  If you have
a paper copy of the Anzac catalog, you can look up the patent numbers
for these amplifiers.  I don't have my catalog with me right now, or
I would list them here.  Possibly, the patents cited in the .pdf
Steve mentioned are the same.  Patents run only 20 years, so I can
guarantee you don't have to worry about them.  The breakthrough here
is that with transformer feedback you can get low noise figure and
high third order intercept simultaneously.  You can't do that with
resistive feedback.

When HP (now Agilent) used to make crystal oscillators, they used
AM-123's as buffer amplifers in their phase noise measurement system.
They have extremely low close-in phase noise because of the feedback.
It is necessary to have very low distortion to avoid upconverting
audio 1/f noise into phase noise.

I am lucky enough to have several AMC-XXX series amplifiers (the
AMC is the packaged version).  I use these for antenna preamps.
They have very high third order intercept and work great.  Someone
at a swap meet who didn't know what he had sold them to me.

If you read the patents cited in the Anzac catalog, they give
pretty much complete details about how to build one of these things,
even mentioning the 2N5109 transistor.

The .pdf that Steve mentioned shows common base amplifiers.  With
the low base resistance of the 2N5109, you have to be careful about
VHF oscillations.  Sometimes you have to add a ferrite bead.  If you
use the right ferrite bead, it will be reactive at HF and resistive
at VHF.  This will kill the oscillation without affecting noise figure.
You can also simply use an inductor and resistor in parallel to do this.

Rick N6RK


_______________________________________________
Topband mailing list
Topband@contesting.com
http://lists.contesting.com/mailman/listinfo/topband

<Prev in Thread] Current Thread [Next in Thread>